摘要
结合模糊推理和神经网络两种方法的优点,从网络的结构、工作过程、学习算法等方面,探讨了一种基于模糊神经网络(FNN)的目标识别方法。通过仿真结果证明,此方法确实可行。
The classical statistical reasoning method is usually adopted in target identification, which needs plentiful prior information. An intelligent method is more effectual, because the target identification is similar to the person's judgment process. In intelligent method, the fuzzy reasoning (FR) and neural network (NN) need little prior information, only the input, output data and certain rules are needed, so they are more applicable for target identification, which is nonlinear and difficult to set up a model. A target identification method based on Fuzzy Neural Network (FNN) is discussed with its network construction, working process and study algorithm. The method combines the advantages of FR and NN. Simulation result shows that this method is feasible.
出处
《电光与控制》
北大核心
2005年第3期50-54,共5页
Electronics Optics & Control
关键词
模糊推理
神经网络
BP学习算法
目标识别
fuzzy reasoning
neural network
back propagation
target identification